The present disclosure describes how to capture events (e.g., changes and user interactions) of a web document and combine those changes with the original DOM displayed to accurately and efficiently enable a replay engine to redisplay the DOM, changes, and user interactions which occurred within a user's browser. The data collected from a client-side HTML DOM capture engine can be combined with a minimal amount of contextual information to a replay engine so as to accurately and efficiently replay a session of a plurality of web documents.
|
1. A method for tracking events associated with a web document on a client device, the method comprising performing, by the client device:
receiving the web document at the client device, the web document having nodes in a document object model (DOM), wherein the DOM is a tree data structure;
receiving a capture agent in conjunction with a delivery of the web document, the capture agent configured to execute on the client device;
capturing, by the capture agent, a plurality of events associated with a plurality of nodes in the DOM, wherein each of the plurality of events includes one or more changes to the DOM, wherein the plurality of events include user interactions on the web document;
for each of the plurality of events:
determining a timestamp for the event;
determining identification information of an associated node, wherein the associated node is associated with the one or more changes to the DOM, and
storing the identification information, the timestamp, and the one or more changes in an event record; and
transmitting, from the capture agent, the event records to a server-side web session storage engine for replaying the user interactions according to the timestamps of the plurality of events.
19. A computer product comprising a computer readable medium storing a plurality of instructions for controlling a client device to track events associated with a web document on the client device, the plurality of instructions comprising:
receiving the web document at the client device, the web document having nodes in a document object model (DOM), wherein the DOM is a tree data structure;
receiving a capture agent in conjunction with a delivery of the web document, the capture agent configured to execute on the client device;
capturing, by the capture agent, a plurality of events associated with a plurality of nodes in the DOM, wherein each of the plurality of events includes one or more changes to the DOM, wherein the plurality of events include user interactions on the web document;
for each of the plurality of events:
determining a timestamp for the event;
determining identification information of an associated node, wherein the associated node is associated with the one or more changes to the DOM, and
storing the identification information, the timestamp, and the one or more changes in an event record; and
transmitting, from the capture agent, the event records to a server-side web session storage engine for replaying the user interactions according to the timestamps of the plurality of events.
2. The method of
searching for a unique DOM node identifier of the associated node.
3. The method of
when a unique DOM node identifier is not found for the associated node:
searching for a sibling or ancestor node that has a unique DOM node identifier;
when the sibling or ancestor node that has the unique DOM node identifier is identified, determining a path from the sibling or ancestor node to the associated node; and
storing the unique DOM node identifier of the sibling or ancestor node and the path in the event record.
4. The method of
5. The method of
when no sibling or ancestor node that has a unique DOM node identifier is identified:
storing DOM element attributes that are unique among sibling DOM elements to identify a first node in the path to the associated node.
6. The method of
when a unique DOM node identifier is not found:
storing DOM element attributes of the associated node that are unique.
7. The method of
8. The method of
receiving the event records at the server-side web session storage engine; and
combining the event records with a server-side captured DOM of the web document to generate a modified DOM from an original unmodified DOM.
9. The method of
10. The method of
identifying, by the capture agent, overlapping modification events; and
storing only one event record for the overlapping modification events.
11. The method of
12. The method of
merging, by the capture agent, adjacent sibling DOM text nodes into a single node for identification of the associated node of an event.
13. The method of
storing, by the capture agent, the event records in client storage until the event records are transmitted to the server-side web session storage engine; and
deleting the event records after sending the event records to the server-side web session storage engine.
14. The method of
subsequent to deleting the event records, capturing a plurality of additional events associated with nodes in the DOM;
for each of the plurality of additional events:
determining additional identification information of an additional associated node, and
storing the additional identification information in an additional event record; and
transmitting, from the capture agent, the additional event records to the server-side web session storage engine.
15. The method of
displaying the web document by a web browser, wherein the capture agent sends the event records when the web browser is not making any other network requests.
16. The method of
compressing, by the capture agent, the event records before transmitting the event records to the server-side web session storage engine.
17. The method of
stripping, by the capture agent, sensitive information before transmitting the event records to the server-side web session storage engine.
18. The method of
|
|||||||||||||||||||||||||||||||||||||||
The present application is a continuation of, and claims the benefit and priority to U.S. application Ser. No. 14/984,102, filed Dec. 30, 2015, entitled “ACCURATE AND EFFICIENT RECORDING OF USER EXPERIENCE, GUI CHANGES AND USER INTERACTION EVENTS ON A REMOTE WEB DOCUMENT”, which claims the benefit and priority under 35 U.S.C. 119(e) of U.S. Provisional Application No. 62/098,951, filed Dec. 31, 2014, the entire contents of which are incorporated herein by reference for all purposes.
The present disclosure relates generally to capturing dynamic real-time changes to a web document within a browser. More specifically, the present disclosure relates to systems and methods for the remote capture of user interaction events and web document or graphical user interface changes within a web document for user experience analysis, playback, and statistical analysis.
Since the introduction of client-side web technologies such as JavaScript, web documents have become increasingly dynamic, allowing users to interact directly with the web document without the browser making time-consuming requests to a server for each content change. Interacting with a data table to sort columns or filter data, changes to the Document Object Model (DOM) element opacity/location/dimensions/content, and asynchronously fetching and displaying data are just some examples of how web documents have become progressively richer with dynamic content and interactions.
Because web document content can be modified within the user's browser, website developers and providers do not have a clear insight into how their audience is using and interacting with the web documents or web applications. Despite having created the web document content, but because of the large amount of permutations of how users can interact with an individual document or groups of documents, providers, designers, operators, and web document creators seek an accurate and efficient way to capture how users interact with their web documents to playback and analyze the remote interactions. Traditionally, the network to desktop browsers was viewed as reliable and adhering to fairly consistent and predictable performance patterns. In the new environment where users increasingly access data from any device, and over widely varying network conditions, the proposition that performance is consistent is no longer valid. Users may see only partial content before getting frustrated and leaving a page or website. They may get frustrated due to content that does not even come from the primary website, but instead is sourced from Content Distribution Networks or external advertising or social media sites.
Products can capture graphical user interface changes on a remote web document. Current capture systems have approached analysis of remote web document events first using log files, and later using either server-side packet capture systems or a client-side capture agent communicating with a server-side storage and analysis system. An advantage exists in combining the server-side packet capture and client-side capture agent, as both capture overlapping data, and the most recent art has not been able to make efficient use of combining the two methods
Attempts to keep a client-side capture agent's data in sync with a replay engine have met various challenges, such as client-side plugins or server-delivered scripts that modify the DOM prior to document load completion and changes to adjacent text nodes resulting in a merging of a node during replay. To compensate for these challenges, systems have been required to send a new copy of the entire DOM or HTML as viewed by the remote browser back to the server-side web session storage and analysis engine, sometimes more than once on a single document to “sync” the replay engine with the actual document being viewed remotely with the client-side capture agent. However, this inefficiency of sending the full HTML from the client both consumes additional client bandwidth and can slow other interactions, for example, the client web browser fetching the next document or new data.
Products can also capture user interaction events on a remote web document. Specifically with respect to capturing mouse clicks and mouse movement, systems have approached tracking of these type of events by capturing the Cartesian coordinates where a mouse moves, and the Cartesian coordinates of where a user performs mouse clicks. This has been moderately effective in the past. However, as the number of devices used to access web documents increase and the number of varying screen displays increases, web designers have transitioned to a more dynamic, or responsive, document design, where the content layout changes dynamically based on the client's screen size. As such, simply capturing Cartesian coordinates is ineffective and inaccurate at helping web operators and designers analyze how users interact with their document, as the results are not clear with respect to what the user clicked or what content a mouse went over, in, or out of.
Therefore, it is desirable to provide new techniques that address these and other problems.
Embodiments provide systems, methods, and apparatuses to accurately and efficiently capture events of a Web document at a client device and send the events to a server-side capture engine, which can combine the events to reassemble the events. For example, user interface and user interaction events on a remotely displayed web document can be reassembled for the purpose of playback and analysis.
In some embodiments, only minimal DOM node modifications are sent without sending the entire DOM tree, while ensuring the DOM model is accurately represented in the server-side storage and analysis system. Additionally, user interaction events such as mouse moving, scrolling, mouse clicks, and keyboard entries can be sent using unique DOM identifiers to accurately playback the events with respect to the content elements that occurred on a remote browser. Embodiments can address many challenges with the complete, yet minimal, capture set required for accurate playback and analysis of user interface and user interaction events, with efficient use of network communication.
Other embodiments are directed to systems and computer readable media associated with methods described herein.
A better understanding of the nature and advantages of embodiments of the present invention may be gained with reference to the following detailed description and the accompanying drawings.
Efficient and accurate capture of user interactions and graphical user interface changes is needed for playback and analysis of web document sessions. There has been a large shift towards access via tablet, mobile, and other non-desktop devices over the last five years. Existing methods do not recognize the unique challenges of capturing and recording user interaction data in this new reality. Embodiments of the present invention have several new innovations that address the challenges of a multi-device reality. In addition, many sites now rely on advertising and/or social media for a substantial portion of the business value. Existing methods also fail to recognize this requirement, and do not provide any insight into how external content is impacting what the user encounters.
Methods and systems for accurate and efficient capture of a web document graphical user interface events (e.g., changes and user interactions) are described. Embodiments can address the inefficiency of sending the entire DOM or HTML by only sending the necessary DOM changes back to the server, and suppressing duplicate information. Embodiments can also address inaccuracies in capturing user interactions by sending specific DOM node identifiers as part of identifying mouse movements and mouse clicks. Some embodiments can replay without any prior configuration or instrumentation of the server-side DOM capture engine.
Embodiments can record everything sent from both sides, both from the server and the client for a specific IP address. One can then pull up the page that the user actually saw (and thus display the pages that the user saw) and replay the steps and interactions the user took (e.g., a stock trade for a financial company). The determination of the actual displayed page can be done even for dynamic web pages. This can be done by listening to those changes and sending the changes back to a server in an efficient manner. Embodiments can link the DOM originally sent by the server with small changes and user interactions on the front end.
I. System
A web session may begin with a user 100 on a web browser 101 initiating the browser request 110 for a web document 150 from a web server 130. As the request is sent to the web server 130, the request may travel through a collective of routers across a network 105. The web server 130 may reside behind a network device 120, such as a load balancer, switch, or router.
In one embodiment, the traffic between the user 100 and web server 130 may be network mirrored 114 to the server-side capture engine 115. The server-side capture engine 115 may exist within the web server 130. In an embodiment where the web server 130 is operating in a cloud environment without administrative control of the network device 120, the web server 130 may copy and forward its packets to the server-side capture engine 115. In another embodiment, the client-side capture agent may send the DOM to the server-side capture engine directly.
Once the web server 130 has received the request, web server 130 may transmit a response 111 through the network device 120 and the network 105 for return to the user 100 web browser 101. The web server 130 may include within the response the HTML Document Object Model (DOM) 151 and an embedded client-side capture agent 152. In one embodiment, a web document can contain a reference to the capture agent (e.g., JavaScript code), which can then be fetched as a result. The capture agent could also be sent separately from the web document. Regardless, the capture agent would be sent in conjunction with the web document, such that the capture agent can capture data about the web document. Thus, the a capture agent can be received in conjunction with a delivery of the web document.
The server-side capture engine 115 stores a copy of the request and response from the web server 130 as 116, which may include the entire DOM 151. The server-side captured data can be stored in the server-side capture engine 115. Engine 115 may include a plurality of separate engines, such as the capture engine, an analysis engine, and a web session storage engine.
The embedded client-side capture agent 152 can monitor changes to the DOM 151 and user 100 interactions, such as mouse clicks, mouse movements, scrolling, and keyboard entry to relay action and changes through the initiation of additional browser requests 110 to the web server 130. Changes to DOM 151 and user interaction are examples of events associated with nodes in the DOM. When these requests return to the web server 130, they are captured by the server-side capture engine 115 and the client-side data is stored as 117, as detailed above for later reassembly and analysis.
Example embodiments of the embedded client-side capture agent 152 are further detailed below. The server-side capture engine and web session storage 115 can combine the data from the client-side agent 117 with the data from the captured server-side network traffic 116 to create an accurate representation of all user interface and user interaction events. The server-side capture engine 115 may use various combinations of IP address, referrer, session cookies, browser identifier, and operating system type to link web documents together into a session for replay.
II. Client-Side Capture
Server-side packet capture systems can capture the HTML/DOM sent to the browser. But, before the browser has completed loading the DOM sent from the server and captured server-side, the HTML may be modified using, perhaps, a client-side plugin or server-delivered scripts. Modern browser technology does not provide a method to distinguish which DOM nodes were modified client-side and which were delivered from the original HTML sent from and captured on the server-side. With this regard, using the full tree DOM path to address changes may lead to errors during replay and analysis when client-side scripts modify the DOM during the DOM load in the browser and prior to the capture agent loading.
To address this challenge, embodiments can determine identification information of a node associated with an event, and save that identification information with an event record. As an example of using identification information, embodiments can direct the capture agent to use the nearest uniquely identified ancestor node to uniquely name the path to the current modification. A uniquely identified node is a node that can be unambiguously identified without using a DOM path, such as an HTML element with an “id” attribute field. An example embodiment of this process is detailed in
At block 200, the DOM is loaded in the browser 200. Then, at block 201, the capture agent is loaded. For example, the browser can begin loading resources, including the HTML document, javascript, images, stylesheets and other data required to render the page. The HTML document parsed by the browser can include instructions to load the capture agent software into the browser.
At block 202, the capture agent begins monitoring for DOM changes (e.g., graphical user interface changes) and user interaction events. While processing changes and events, the capture agent may use memory available within the browser, or a permanent storage mechanism provided by the browser. Some events, such as mouse movements, may be sampled to a specific time resolution, such as 0.1 seconds, to reduce the amount of data collected. Other events, such as a mouse click, may not be sampled to ensure an accurate representation of the user's interactions.
At block 210, when the capture agent receives a DOM modification notification event, the capture agent begins to process the changes. The DOM modification notification event may contain multiple DOM changes. As examples, the DOM modification can include an addition of one or more nodes in the DOM, a removal of one or more nodes in the DOM, or a modification to one or more existing nodes in the DOM.
At block 211, to efficiently process the changes within the modification notification event, the capture agent ignores children node modifications for ancestor nodes which were modified and which required the DOM content to be sent to the server-side capture engine. An example embodiment of this process is detailed in
At block 212, the capture agent looks for adjacent text nodes in additions, modifications, or removals that can be represented as one text node during replay. Example embodiments of these processes are detailed below in
At block 213, the capture agent uses nearest uniquely identified ancestor DOM element to identify a path to the event. If the associated node has a unique identification, then an ancestor is not needed. Some embodiments can unambiguously identify a target node of DOM changes by constructing a sequence of node indexes by recursively taking the node index of the target node, then the node index of the target node's parent and so on until a node is reached that can be unambiguously identified through another means, such as the fact that the node is the root node of the entire DOM. In this description, this sequence of node indexes is called a “DOM path”. A DOM path that ends with the root node of the entire DOM is called a “full tree DOM path”.
At block 230, once the capture agent has identified modification(s) to send to the server-side capture, storage, and analysis engine, the capture agent may strip out sensitive information. The sensitive information may include elements such as passwords, credit cards, social security numbers, and other configured sensitive fields. Thus, the capture agent can strip sensitive information before transmitting the event records to the server-side web session storage engine.
At block 231, the capture agent may then assemble and coalesce the data, including meta information such as processing time, load time, and other situational information available within the browser, timestamps, and other information such as errors. It may coalesce the data (e.g., over a period of 5 seconds) by combining similar events and using short identifiers to optimize the information into the smallest amount of data bytes.
At block 232, the capture agent may compress the data in pre-determined chunks. Various compression algorithms are known to those skilled in the art, which trade time to compress vs compression efficiency. While one compression algorithm may be applicable today given today's CPU processing power available, in the future, a more time consuming algorithm may be more appropriate.
At block 233, the capture agent may send the data to the server-side storage and analysis engine. The server can use the data (e.g., as event records) to replay the changes. For example, the server can identify an event and the corresponding node. The server can then make that change and replay it. After receiving the event records at the server-side web session storage engine, the server-side web session storage engine can combine the event records with a server-side captured DOM of the web document to generate a modified DOM from an original unmodified DOM.
In some embodiments, the capture agent can store the event records in a client storage until the event records are transmitted to the server-side storage and analysis engine. The event records can be deleted after sending the event records to the server-side engine. Subsequent to deleting the event records, a plurality of additional events associated with nodes in the DOM can be captured. For each of the plurality of additional events, additional identification information of an additional associated node can be determined. The additional identification information can be stored in an additional event record. The capture agent can transmit the additional event records to the server-side web session storage engine.
In one embodiment, the client-side capture agent sends the data at pre-determined intervals or when the document is unloaded in the browser. The data can be sent at pre-determined intervals to ensure data is not lost in the browser or during transit. As another example, the client-side capture agent can send the data when there are no existing network requests.
At block 220, the capture agent may also receive user interaction events, such as keyboard entry, document resize, orientation changes, scroll events, mouse movements, or mouse clicks. In the case of keyboard entry, the capture agent may strip sensitive information from the data (event record) as detailed above. The user interaction data can follow a similar path as for DOM modifications detailed above in block 213 for the capture agent identifying the DOM node of the target of the events, in block 230 stripping sensitive information, in block 231 coalescing the data, in block 232 compressing the data, and in block 233 sending the data.
As examples, four different ways can be used in order to determine a unique identification. DOM node identifiers can be used. Embodiments may potentially use sibling or ancestor node identifiers. If the current node is not uniquely identified, one can go up to parents. Once embodiments start looking up at the parents, embodiments can look at siblings. Embodiments can look for a unique parent or sibling, such as the case where there is a parent that is, for example, a DIV (division in HTML) or a P (paragraph in HTML). And then inside the paragraph there are three spans. If none of them had unique identifications on them (e.g., no unique identifier), the DIV may have a unique identifier. But, if there is a change to the third span, embodiments can look to see that there are three span children and that current node is a third span child.
This can use both sibling or ancestor DOM node identifiers and DOM element attributes that are unique. This can use the order of the node among its siblings. Embodiments can also find DOM element attributes that are unique among sibling DOMs. For example, one span can have a text size of 25 and it is being modified to text size of 22, so that is how embodiments can uniquely identify it. When no sibling or ancestor node that has a unique DOM node identifier is identified, DOM element attributes that are unique among sibling DOM elements can be stored to identify a first node in the path to the associated node.
In one example with JavaScript, one might call out a specific identifier (e.g., a book) in JQuery. There might be a number of pages that are children in the book, and so a user might request the element of book and add a new child. So in other words, a user wants to add a new page. Thus, there are use cases where embodiments might address things where there is a unique identifier like book, the book tag or element, or node, and then address it from a perspective of children.
III. Uniquely Identifying a Node in a DOM Tree
A. Path to a Node in DOM Tree
At block 300, the capture agent receives the DOM modifications notification. The notification can be received in any suitable form, e.g., as a flag, a flag with a message, a message that include information about the modification, etc.
At block 310, when the capture agent receives the DOM modifications notification, the capture agent evaluates if the action is a node addition or removal. The DOM modification event can include indicators if the node is being added or removed.
At block 320, the capture agent may then look for a unique identifier on the DOM node, such as an ‘id’ attribute which would uniquely identify the node. Another example is combinations of attributes on a specific tag name, such as input[name=“address1”], which would uniquely identify the input field for the address 1.
At block 321, the capture agent may begin to traverse the tree upwards looking for a unique identifier. The traversal may occur in any suitable order. For example, the tree can be traversed via each branch until an end is reached, and then a previous branch point not taken can be traversed.
At block 330, to identify the unique tree path to the modified node, the agent uses the path to the node changed with a root of the node with unique ID to represent the unique tree path. The path may be represented by node indexes such as those illustrated in
B. Node Index
To send and store DOM changes, the nodes to which the changes occur (i.e. the targets of the changes) can be unambiguously identified. Because each node in the DOM has a well-ordered set of child nodes, this can be accomplished by defining the node index of a node to be the ordinal number of the node in the well-ordered set of child nodes of the node's parent node. An example embodiment of the process of calculating the node index is detailed in FIGS. 4A and 4B. This can also be accomplished by defining the node index of a node to be the ordinal number of the node in the well-ordered set of child nodes, filtered for only child nodes of the same type as the target node, or of the node's parent node. An example embodiment of the process of calculating the node index for nodes of the same type is detailed in
At block 400, a counter is set to zero in the client-side agent. The counter can be stored in any suitable memory and associated with a process for calculating a node index.
At block 410, for the purposes of iteratively looping through the DOM tree to unambiguously identify the location of the node, a variable, or current node, is set to the first child of the target node's parent node, where the target node is the node to unambiguously identify.
At block 420, it is determined whether the current node is the target node, e.g., by comparing a reference, or memory location, of the node being evaluated to the reference, or memory location, of the node to unambiguously identify.
At block 421, the counter is incremented by one if the current node is not the target node. On the next iteration, the current node is changed to a next child of the target node's parent node.
At block 430, the final value of the counter is the node index. This index is used to create an unambiguous path from a uniquely identified node to the target node. Using
At block 450, a counter is set to zero in the client-side agent. The counter can be stored in any suitable memory and associated with a process for calculating a node index.
At block 460, for the purposes of iteratively looping through the DOM tree to unambiguously identify the location of the node, a variable, or current node, is set to the first child of the target node's parent node, where the target node is the node to unambiguously identify.
At block 470, it is determined whether the current node is the target node.
At block 471, it is determined whether the current node is a same node type as the target node. Two nodes share the same type when they have the same node tag. For example, to nodes would be the same type if they were both “<span>” nodes.
At block 472, if the current node is not the target node 470, and if the current node is the same type of node as the target node 471, the counter is incremented by one.
At block 480, if the node is the target node 470, the final value of the counter is the node index.
IV. Overlapping Modifications
At block 500, when the capture agent receives the DOM modifications notification, the notification may contain multiple modifications. For the purposes of efficiently, modern day browsers currently send DOM modifications in groups of changes. The capture agent analyzes the set of changes to suppresses duplicate information from overlapping modifications.
At block 510, the capture agent evaluates if the notification event contains multiple events, e.g., by examining the length of the array of changes.
At block 520, if the capture agent does contain multiple events, the capture agent creates a first list of modified nodes and a second list of parent nodes that has had a child added or a child removed. Thus, the modified nodes of the first list already existed and attributes of the node itself have been modified. The nodes of the second list have had a child node added or removed. It is possible that nodes can be in both lists, in the case where a node is added and then removed. In this case, the capture agent will ignore the node modification altogether as the end status of the node is removed.
At block 530, the capture agent iterates over the first list of modified nodes.
At block 540, at each iteration, the capture agent checks if an ancestor of the current node is in the second list of parent nodes with a child added or removed.
At block 560, if the ancestor of the current node is in the list of parent nodes with a child added or removed from block 540, the current modification may be ignored by the capture agent, as this modification may already be sent with the ancestor changes.
At block 550, if the current node does not have an ancestor in the list of parent nodes with a child added or removed from block 540, or if the notification event did not contain multiple events from block 510, then the modification data is prepared to later send to the server-side storage and analysis engine.
At block 570, the capture agent continues to iterate over all the modified nodes until complete. Accordingly, the capture agent can identify overlapping modification events, and store only one event record for the overlapping modification events. In one embodiment, a modification of a single ancestor node can represent overlapping modification events targeting a single ancestor node subtree. For instance, overlapping modifications can be within a single parent node that has more than one child or extended grandchildren that has been modified (and thus overlapping with the modification of the ancestor which must be serialized). An example is when A has two children B and C. B has one child D. C has two children F and G. F has one child H. H has three children I, J, and K. If there is a modification to J, G, and C, embodiments only need to send the modification for C as it would contain all modification of C (i.e., including any modification of F, G, H, I, J, and K), and thus there is no need to record J and G changes.
V. Merging Text
To send and store DOM changes, the changes can be serialized. In one embodiment, the serialization of a node being added may include the identifier of the parent of the target node as obtained by the process detailed above, the node index of the new node, and the HTML representation of the node and its children. Obtaining the HTML representation of a DOM node can be done in a computationally efficient manner as this is a native feature in modern browsers and is performed with machine code rather than JavaScript. The drawback is that this HTML representation is ambiguous about whether contiguous text is represented as one text node in the DOM (the “canonical” form of the DOM) or as multiple text nodes. DOM modification as a result of user interaction or user interface changes may cause multiple sibling text nodes to appear through the addition and modification of text nodes. These adjacent nodes are represented as a single node during replay and can lead to errors if not properly managed during capture.
To address this challenge, changes are serialized in such a manner so that an agent replaying or analyzing the serialized changes may assume that the beginning state of any set of changes is a canonical DOM: that is, there are no adjacent text nodes being that all of them are merged together. Note that the serialized changes themselves may render the DOM non-canonical, but it is assumed that the DOM is changed to return to a canonical state before the next set of serialized changes. This may be affected with three additions to the way DOM changes are serialized:
When text nodes are modified, if the modification's target is a text node that has adjacent text nodes, a modification to a “virtual text node” that combines the adjacent text nodes into one text node positioned at the beginning of the sequence of contiguous text nodes is serialized instead of a modification to the actual target text node. An example embodiment of this process is detailed in
For purposes of calculating node indexes, adjacent text nodes are counted as one text node, the canonical node. An example embodiment of the process of determining the canonical node index is detailed in
For the addition of any node between two adjacent text nodes, the capture agent can also serialize a modification to the DOM such that the virtual text node combining the two adjacent text nodes are split into two or more text nodes, with at least one of the breaks occurring where the new DOM node is added. In one embodiment, a text node modification may be serialized for the virtual text node that removes the text that belongs to the text nodes appearing after the DOM node being added. A DOM node addition may be serialized to add a text node with the previously removed text after the virtual text node. Finally, a DOM node addition can be serialized for the DOM node being added between the two text nodes. An example embodiment of this process is detailed in
This section particularly relates to the way embodiments can serialize additions of DOM nodes as the HTML of the node and its children. A possible drawback is that this serialization may be ambiguous about whether contiguous text in the serialization is represented as one text node in the DOM or as multiple text nodes. A solution is to always serialize changes such that the beginning state of any change we serialize is a canonical DOM (that is, there are no adjacent text nodes, all of them being merged together).
At block 600, the capture agent detects a text node modification. For example, the capture agent can review the node types declared on each node in the list of modifications to determine if any of the nodes are of type text node.
At block 610, the capture agent determines if there are sibling text nodes. If so, the capture agent may serialize the modification as if it occurred to a single virtual text node that combines the contiguous sequence of text nodes including the text node for which the change was detected. The modification can be depicted as positioned at the start of the contiguous sequence of text nodes.
At block 620, the capture agent finds the beginning of the contiguous sequence. The beginning of the sequence is the first node in the ordered list of nodes that is of type text node.
At block 630, the capture agent concatenates all the text in the contiguous sequence.
At block 640, the capture agent then serializes the modification for the position determined in block 620 and the text determined in block 630. Accordingly, the capture agent can merge adjacent sibling DOM text nodes into a single node for identification of the associated node of an event. In one embodiment, the serialization may be a text representation of the text of the modified node. In another embodiment, serialization may be to a structure stream of data which may include other attributes of the text node to ensure fidelity of the all of the node's data.
At block 650, if there were no sibling text nodes, the capture agent serializes the modification for the position and text of the target text node.
At block 700, a counter is set to zero in the client-side agent.
At block 710, for the purposes of iteratively looping through the DOM tree to unambiguously identify the location of the node, the current node is set to the first child of the target node's parent node, where the target node is the node to unambiguously identify.
At block 720, a loop is created while the current node is not the target node.
At block 721, it is determined whether the current node is a text node. If the current node is not a text node, the counter is incremented by one at block 723. If the current node is not a text node, the method proceeds to block 722.
At block 722, it is determined whether the node previous to the current node is a text node. If the previous node is not a text node, the counter is incremented by one at block 723
At block 730, the final value of the counter is the node index.
At block 810, after the capture agent detects a DOM node addition modification 800, the capture agent may evaluate if it is between two adjacent text nodes. For example, if the node has sibling nodes, the capture agent can check if either of the adjacent nodes is of type text node.
At block 820, if the modification is between two adjacent text nodes, the capture agent can find the beginning of the contiguous sequence of text nodes that contains the text node preceding the addition. To do so, the capture agent can iterate through the list of sibling nodes, checking for type of text node. The iteration in either direction can stop once a non-text node is reached or the end of the sibling list in that direction is reached.
At block 830, the capture agent can concatenate the text of the contiguous sequence of text nodes that starts at the determining beginning of the contiguous sequence.
At block 840, the capture agent may then serialize the text from the position determined in 820 and the text determined in 830. This text represents the entirety of the text node that will be seen from the replay's perspective, where all contiguous text nodes will be a single text node. An example would be where the following nodes exist:
<div>
Welcome to
<br>
Quantum Metric
<br>
2015
</div>
If a modification occurred removing the two<br> nodes, the “Welcome to”, “Quantum Metric”, and “2015” nodes would continue to be adjacent nodes during the capture, and they would continue to be represented as 3 distinct text nodes. During replay, these 3 text nodes would be a single text node “Welcome to Quantum Metric 2015”.
At block 850, the capture agent may then concatenate the text of the contiguous sequence of text nodes starting with the text node following the addition.
At block 860, the capture agent may then serialize a node addition to the text representation of the text contained in the text node, for a text node at the position of the text node following the addition with the text determined in 850.
At block 870, if the modification is not between two adjacent text nodes, the capture agent may then serialize a node addition for the node to be added, where in contrast to when other text nodes are adjacent, the text node is treated similarly to other node additions.
Any of the computer systems mentioned herein may utilize any suitable number of subsystems. Examples of such subsystems are shown in
The subsystems shown in
A computer system can include a plurality of the same components or subsystems, e.g., connected together by external interface 81 or by an internal interface. In some embodiments, computer systems, subsystem, or apparatuses can communicate over a network. In such instances, one computer can be considered a client and another computer a server, where each can be part of a same computer system. A client and a server can each include multiple systems, subsystems, or components.
It should be understood that any of the embodiments of the present invention can be implemented in the form of control logic using hardware (e.g. an application specific integrated circuit or field programmable gate array) and/or using computer software with a generally programmable processor in a modular or integrated manner. As used herein, a processor includes a single-core processor, multi-core processor on a same integrated chip, or multiple processing units on a single circuit board or networked. Based on the disclosure and teachings provided herein, a person of ordinary skill in the art will know and appreciate other ways and/or methods to implement embodiments of the present invention using hardware and a combination of hardware and software.
Any of the software components or functions described in this application may be implemented as software code to be executed by a processor using any suitable computer language such as, for example, Java, C, C++, C #, Objective-C, Swift, or scripting language such as Perl or Python using, for example, conventional or object-oriented techniques. The software code may be stored as a series of instructions or commands on a computer readable medium for storage and/or transmission, suitable media include random access memory (RAM), a read only memory (ROM), a magnetic medium such as a hard-drive or a floppy disk, or an optical medium such as a compact disk (CD) or DVD (digital versatile disk), flash memory, and the like. The computer readable medium may be any combination of such storage or transmission devices.
Such programs may also be encoded and transmitted using carrier signals adapted for transmission via wired, optical, and/or wireless networks conforming to a variety of protocols, including the Internet. As such, a computer readable medium according to an embodiment of the present invention may be created using a data signal encoded with such programs. Computer readable media encoded with the program code may be packaged with a compatible device or provided separately from other devices (e.g., via Internet download). Any such computer readable medium may reside on or within a single computer product (e.g. a hard drive, a CD, or an entire computer system), and may be present on or within different computer products within a system or network. A computer system may include a monitor, printer, or other suitable display for providing any of the results mentioned herein to a user.
Any of the methods described herein may be totally or partially performed with a computer system including one or more processors, which can be configured to perform the steps. Thus, embodiments can be directed to computer systems configured to perform the steps of any of the methods described herein, potentially with different components performing a respective steps or a respective group of steps. Although presented as numbered steps, steps of methods herein can be performed at a same time or in a different order. Additionally, portions of these steps may be used with portions of other steps from other methods. Also, all or portions of a step may be optional. Additionally, any of the steps of any of the methods can be performed with modules, circuits, or other means for performing these steps.
The features and advantages described in the detailed description are not all inclusive and, in particular, many additional features and advantages will be apparent to one of ordinary skill in the art in view of the drawings, detailed description, and claims. Moreover, it should be noted that the language used in the detailed description has been principally selected for readability and instructional purposes, and may not have been selected to delineate or circumscribe the inventive subject matter.
Note that in this description, references to “one embodiment,” “an embodiment” or “some embodiments” mean that the feature being referred to is included in at least one embodiment of the invention. Further, separate references to “one embodiment” or “some embodiments” in this description do not necessarily refer to the same embodiment(s); however, neither are such embodiments mutually exclusive, unless so stated and except as will be readily apparent to those skilled in the art. Thus, the invention can include any variety of combinations and/or integrations of the embodiments described herein. However, other embodiments of the invention may be directed to specific embodiments relating to each individual aspect, or specific combinations of these individual aspects.
Upon reading this detailed description, those of skill in the art will appreciate still additional alternative structural and functional designs for a system and method for accurate and efficient capture of user interface and user interaction events on a remote web document through the disclosed principles of the present invention. Thus, while particular embodiments and applications of the present invention have been illustrated and described, it is to be understood that the invention is not limited to the precise construction and components disclosed herein and that various modifications, changes and variations which will be apparent to those skilled in the art may be made in the arrangement, operation and details of the method and apparatus of the present invention disclosed herein without departing from the spirit and scope of the invention as defined in the appended claims.
A recitation of “a”, “an” or “the” is intended to mean “one or more” unless specifically indicated to the contrary. The use of “or” is intended to mean an “inclusive or,” and not an “exclusive or” unless specifically indicated to the contrary.
All patents, patent applications, publications, and descriptions mentioned herein are incorporated by reference in their entirety for all purposes. None is admitted to be prior art.
Ciabarra, Jr., Mario Luciano, Wang, Yiduo
| Patent | Priority | Assignee | Title |
| 11487931, | Oct 18 2021 | International Business Machines Corporation | Replaying a webpage based on virtual document object model |
| 11636172, | Dec 31 2014 | Quantum Metric, Inc. | Accurate and efficient recording of user experience, GUI changes and user interaction events on a remote web document |
| 11995145, | Dec 31 2014 | Quantum Metric, Inc. | Accurate and efficient recording of user experience, GUI changes and user interaction events on a remote web document |
| Patent | Priority | Assignee | Title |
| 10063645, | May 10 2012 | Content Square Israel Ltd | Method and system for monitoring and tracking browsing activity on handled devices |
| 10079737, | Sep 13 2013 | Content Square Israel Ltd | Method and system for generating comparable visual maps for browsing activity analysis |
| 10146752, | Dec 31 2014 | QUANTUM METRIC, INC | Accurate and efficient recording of user experience, GUI changes and user interaction events on a remote web document |
| 10318592, | Jul 16 2015 | QUANTUM METRIC, INC | Document capture using client-based delta encoding with server |
| 10599753, | Nov 11 2013 | Amazon Technologies, Inc | Document version control in collaborative environment |
| 10691877, | Feb 07 2014 | Amazon Technologies, Inc | Homogenous insertion of interactions into documents |
| 5765173, | Jan 11 1996 | AUTONOMY, INC | High performance backup via selective file saving which can perform incremental backups and exclude files and uses a changed block signature list |
| 5794254, | Dec 03 1996 | CDD TECHNOLOGIES, LLC | Incremental computer file backup using a two-step comparison of first two characters in the block and a signature with pre-stored character and signature sets |
| 6012087, | Jan 14 1997 | NETMIND TECHNOLOGIES, INC | Unique-change detection of dynamic web pages using history tables of signatures |
| 7269784, | Jan 22 2001 | DIGITAL RIVER, INC | Server-originated differential caching |
| 7296051, | Feb 19 2002 | DIGITAL RIVER, INC | Predictive predownload of templates with delta encoding |
| 7383496, | May 30 2003 | International Business Machines Corporation | Information processing apparatus, program, and recording medium |
| 7437364, | Jun 30 2004 | GOOGLE LLC | System and method of accessing a document efficiently through multi-tier web caching |
| 7523191, | Jun 02 2000 | R2 SOLUTIONS LLC | System and method for monitoring user interaction with web pages |
| 7565423, | Jun 30 2004 | GOOGLE LLC | System and method of accessing a document efficiently through multi-tier web caching |
| 7587398, | Jun 30 2004 | GOOGLE LLC | System and method of accessing a document efficiently through multi-tier web caching |
| 7640262, | Jun 30 2006 | EMC IP HOLDING COMPANY LLC | Positional allocation |
| 7657517, | Mar 16 2001 | JPMORGAN CHASE BANK, N A , AS SUCCESSOR AGENT | Server for synchronization of files |
| 7668880, | May 15 2007 | Offsite computer file backup system providing rapid recovery and method thereof | |
| 7734826, | Mar 16 2001 | JPMORGAN CHASE BANK, N A , AS SUCCESSOR AGENT | Client-server model for synchronization of files |
| 7765274, | Aug 06 2001 | Digital River, Inc. | Differential caching with template, sub-template, and delta information |
| 7844580, | Mar 28 2003 | ORACLE INTERNATIONL CORPORATION | Methods and systems for file replication utilizing differences between versions of files |
| 7941525, | Apr 01 2006 | Content Square Israel Ltd | Method and system for monitoring an activity of a user |
| 7979413, | May 30 2008 | New York University | Automatic generation of embedded signatures for duplicate detection on a public network |
| 8112496, | Sep 24 2004 | Microsoft Technology Licensing, LLC | Efficient algorithm for finding candidate objects for remote differential compression |
| 8224964, | Jun 30 2004 | GOOGLE LLC | System and method of accessing a document efficiently through multi-tier web caching |
| 8230510, | Oct 02 2008 | TREND MICRO INCORPORATED | Scanning computer data for malicious codes using a remote server computer |
| 8266245, | Oct 17 2011 | GOOGLE LLC | Systems and methods for incremental loading of collaboratively generated presentations |
| 8381302, | Sep 14 2009 | CA, INC | Systems and methods for translating non-comparable values into comparable values for use in heuristics |
| 8423616, | May 03 2007 | Microsoft Technology Licensing, LLC | Identifying and correlating electronic mail messages |
| 8595187, | Feb 13 2013 | AUTODESK, Inc | Serialization for delta encoding |
| 8661428, | Apr 25 2008 | VMware LLC | Updating a file using differences and file format therefor |
| 8826033, | Dec 22 2009 | EMC IP HOLDING COMPANY LLC | Data protection using virtual-machine-specific stable system values |
| 8868533, | Jun 30 2006 | ACOUSTIC, L P | Method and apparatus for intelligent capture of document object model events |
| 8914324, | Oct 16 2009 | Veritas Technologies LLC | De-duplication storage system with improved reference update efficiency |
| 8959332, | Dec 23 2009 | International Business Machines Corporation | Deduplication of encrypted data |
| 9112826, | Dec 23 2011 | AKAMAI TECHNOLOGIES, INC | Data differencing across peers in an overlay network |
| 9153239, | Mar 14 2013 | GOOGLE LLC | Differentiating between near identical versions of a song |
| 9471285, | Jul 09 2015 | SOFTWARE INTEGRITY GROUP, INC | Identifying software components in a software codebase |
| 9508081, | Apr 01 2006 | Content Square Israel Ltd | Method and system for monitoring an activity of a user |
| 9633062, | Apr 29 2013 | Amazon Technologies, Inc | Document fingerprints and templates |
| 9684668, | Sep 15 2011 | Veritas Technologies LLC | Systems and methods for performing lookups on distributed deduplicated data systems |
| 9792365, | Dec 31 2013 | Content Square Israel Ltd | Method and system for tracking and gathering multivariate testing data |
| 9917894, | Aug 06 2014 | QUEST SOFTWARE INC F K A DELL SOFTWARE INC ; Aventail LLC | Accelerating transfer protocols |
| 9946724, | Mar 31 2014 | EMC IP HOLDING COMPANY LLC | Scalable post-process deduplication |
| 9959577, | Sep 28 2006 | INTUIT INC.; INTUIT, INC | Tax return preparation automatic document capture and parsing system |
| 9971940, | Aug 10 2015 | GOOGLE LLC | Automatic learning of a video matching system |
| 20020013825, | |||
| 20020138511, | |||
| 20020174180, | |||
| 20030046260, | |||
| 20030172066, | |||
| 20030200207, | |||
| 20040025171, | |||
| 20040162885, | |||
| 20040243936, | |||
| 20040249793, | |||
| 20040267726, | |||
| 20050066037, | |||
| 20050203851, | |||
| 20050262167, | |||
| 20050273592, | |||
| 20050283500, | |||
| 20060112150, | |||
| 20070130188, | |||
| 20070226510, | |||
| 20070288533, | |||
| 20080033913, | |||
| 20080091845, | |||
| 20080208979, | |||
| 20080235200, | |||
| 20080243898, | |||
| 20080243992, | |||
| 20080270436, | |||
| 20090012984, | |||
| 20090013414, | |||
| 20090177959, | |||
| 20090196296, | |||
| 20090228680, | |||
| 20090248793, | |||
| 20090271779, | |||
| 20090299994, | |||
| 20100088551, | |||
| 20100222902, | |||
| 20110066628, | |||
| 20110119327, | |||
| 20110252100, | |||
| 20110252305, | |||
| 20110314070, | |||
| 20110320880, | |||
| 20120016882, | |||
| 20120060082, | |||
| 20120084333, | |||
| 20130013618, | |||
| 20130013859, | |||
| 20130031056, | |||
| 20130124472, | |||
| 20130138620, | |||
| 20130138775, | |||
| 20130185387, | |||
| 20130188926, | |||
| 20130232187, | |||
| 20130238730, | |||
| 20130262567, | |||
| 20130275479, | |||
| 20130346374, | |||
| 20140032513, | |||
| 20140074783, | |||
| 20140181034, | |||
| 20140259157, | |||
| 20140358938, | |||
| 20140379823, | |||
| 20150033120, | |||
| 20150067031, | |||
| 20150134669, | |||
| 20150142756, | |||
| 20150181269, | |||
| 20150207837, | |||
| 20150261653, | |||
| 20150286511, | |||
| 20150356116, | |||
| 20160055196, | |||
| 20160188411, | |||
| 20160188548, | |||
| 20160205221, | |||
| 20160308941, | |||
| 20160320945, | |||
| 20170011049, | |||
| 20170017650, | |||
| 20170192876, | |||
| CA2437281, | |||
| EP3323053, | |||
| WO2075539, |
| Executed on | Assignor | Assignee | Conveyance | Frame | Reel | Doc |
| Dec 28 2015 | CIABARRA, MARIO LUCIANO, JR | Quantum Metric, LLC | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 048157 | /0046 | |
| Dec 29 2015 | WANG, YIDUO | Quantum Metric, LLC | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 048157 | /0046 | |
| Jul 16 2017 | Quantum Metric, LLC | QUANTUM METRIC, INC | CHANGE OF NAME SEE DOCUMENT FOR DETAILS | 056585 | /0551 | |
| Nov 30 2018 | Quantum Metric, Inc. | (assignment on the face of the patent) | / | |||
| Oct 05 2023 | QUANTUM METRIC, INC | JPMORGAN CHASE BANK, N A | SECURITY INTEREST SEE DOCUMENT FOR DETAILS | 065138 | /0586 |
| Date | Maintenance Fee Events |
| Nov 30 2018 | BIG: Entity status set to Undiscounted (note the period is included in the code). |
| Dec 20 2018 | SMAL: Entity status set to Small. |
| Date | Maintenance Schedule |
| Jun 15 2024 | 4 years fee payment window open |
| Dec 15 2024 | 6 months grace period start (w surcharge) |
| Jun 15 2025 | patent expiry (for year 4) |
| Jun 15 2027 | 2 years to revive unintentionally abandoned end. (for year 4) |
| Jun 15 2028 | 8 years fee payment window open |
| Dec 15 2028 | 6 months grace period start (w surcharge) |
| Jun 15 2029 | patent expiry (for year 8) |
| Jun 15 2031 | 2 years to revive unintentionally abandoned end. (for year 8) |
| Jun 15 2032 | 12 years fee payment window open |
| Dec 15 2032 | 6 months grace period start (w surcharge) |
| Jun 15 2033 | patent expiry (for year 12) |
| Jun 15 2035 | 2 years to revive unintentionally abandoned end. (for year 12) |